Routing scorecard

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description adds significant behavioral context beyond the readOnlyHint annotation: it states that tiers with fewer than minSamples get 'insufficient data' instead of a verdict, that includeRows can blow the result cap, and that the tool never writes model:/cap labels. These details fully disclose the tool's behavior and limitations, aligning with and enriching the annotations without contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but densely informative, with a clear structure: the core question first, then data sources, metrics, cross-cutting by effort, caveats about insufficient data, read-only reassurance, sibling pairing, and return-value overview. Every sentence contributes meaningful detail without fluff.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (5 parameters, multiple metrics, conditional behavior) and no output schema, the description fully covers what the tool returns (per-tier and per-effort stats, opt-in rows), how it behaves with insufficient data, and the read-only nature. It provides a complete mental model for correct invocation and interpretation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema covers 80% of parameters with descriptions, and the description adds practical meaning: it explains the minSamples threshold behavior, the includeRows opt-in and its result-cap impact, and the windowDays rationale ('older model generation is worse than none'). This goes beyond simple schema repetition, providing actionable guidance for parameter usage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states what the tool does: it scores every Done ticket from board data and reports per-tier metrics like cost per clean ticket. It distinguishes itself from siblings by explicitly mentioning pairing with plan_budget and daily_plan, and by framing its purpose as measuring actual performance rather than guessing.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides clear context on when to use this tool: for measuring which model tier actually runs tickets, and explicitly advises pairing it with plan_budget (what the queue will cost) and daily_plan (what to run today). It also notes the tool is 'advice only' and never writes labels, implying it is for analysis rather than action, but it does not explicitly state exclusions like 'do not use for modifying tiers'.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.